Runtime features

Incentivizing to create more practical models


Submit a model from SIGNATE Runtime and measure its inference time.




Submitting a model


In Runtime-supported competitions, participants are asked to submit both a pre-trained model itself and a script to execute an inference, instead of a prediction file. Once a submission is made successfully, a prediction file will be generated automatically and its evaluation result will be notified.


* There is a coding conventions such as method names to submit a model. Please click here for further details.





A high-performance model: Fast and accurate


By submitting a model, the model's inference time (time taken to generate a prediction file) is measured to be displayed on the competition leaderboard along with the score.

In real-world applications, there is a need for models not only with high accuracy but also with high-speed inference. Create more practical model and astonish other participants.


* The model inference time will be notified via an email along with its score.




Submitting from SIGNATE Runtime


In order to submit a pre-trained model from Runtime, a submission file should be created under some conventions such as folder structure and method names.

Click the button below for further information


See Runtime submission guidelines

Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select at least one of the reasons
CAPTCHA verification is required.

Feedback sent

We appreciate your effort and will try to fix the article